You’re out of free articles.
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Sign In or Create an Account.
By continuing, you agree to the Terms of Service and acknowledge our Privacy Policy
Welcome to Heatmap
Thank you for registering with Heatmap. Climate change is one of the greatest challenges of our lives, a force reshaping our economy, our politics, and our culture. We hope to be your trusted, friendly, and insightful guide to that transformation. Please enjoy your free articles. You can check your profile here .
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Subscribe to get unlimited Access
Hey, you are out of free articles but you are only a few clicks away from full access. Subscribe below and take advantage of our introductory offer.
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Create Your Account
Please Enter Your Password
Forgot your password?
Please enter the email address you use for your account so we can send you a link to reset your password:
AI has already changed weather forecasting forever.
It’s been a wild few years in the typically tedious world of weather predictions. For decades, forecasts have been improving at a slow and steady pace — the standard metric is that every decade of development leads to a one-day improvement in lead time. So today, our four-day forecasts are about as accurate as a one-day forecast was 30 years ago. Whoop-de-do.
Now thanks to advances in (you guessed it) artificial intelligence, things are moving much more rapidly. AI-based weather models from tech giants such as Google DeepMind, Huawei, and Nvidia are now consistently beating the standard physics-based models for the first time. And it’s not just the big names getting into the game — earlier this year, the 27-person team at Palo Alto-based startup Windborne one-upped DeepMind to become the world’s most accurate weather forecaster.
“What we’ve seen for some metrics is just the deployment of an AI-based emulator can gain us a day in lead time relative to traditional models,” Daryl Kleist, who works on weather model development at the National Oceanic and Atmospheric Administration, told me. That is, today’s two-day forecast could be as accurate as last year’s one-day forecast.
All weather models start by taking in data about current weather conditions. But from there, how they make predictions varies wildly. Traditional weather models like the ones NOAA and the European Centre for Medium-Range Weather Forecasts use rely on complex atmospheric equations based on the laws of physics to predict future weather patterns. AI models, on the other hand, are trained on decades of prior weather data, using the past to predict what will come next.
Kleist told me he certainly saw AI-based weather forecasting coming, but the speed at which it’s arriving and the degree to which these models are improving has been head-spinning. “There's papers coming out in preprints almost on a bi-weekly basis. And the amount of skill they've been able to gain by fine tuning these things and taking it a step further has been shocking, frankly,” he told me.
So what changed? As the world has seen with the advent of large language models like ChatGPT, AI architecture has gotten much more powerful, period. The weather models themselves are also in a cycle of continuous improvement — as more open source weather data becomes available, models can be retrained. Plus, the cost of computing power has come way down, making it possible for a small company like Windborne to train its industry-leading model.
Founded by a team of Stanford students and graduates in 2019, Windborne used off-the-shelf Nvidia gaming GPUs to train its AI model, called WeatherMesh — something the company’s CEO and co-founder, John Dean, told me wouldn’t have been possible five years ago. The company also operates its own fleet of advanced weather balloons, which gather data from traditionally difficult-to-access areas.
Standard weather balloons without onboard navigation typically ascend too high, overinflate, and pop within a matter of hours (thus becoming environmental waste, sad!). Since it’s expensive to do launches at sea or in areas without much infrastructure, there’s vast expanses of the globe where most balloons aren’t gathering any data at all.
Satellites can help, of course. But because they’re so far away, they can’t provide the same degree of fidelity. With modern electronics, though, Windborne found it could create a balloon that autonomously changes altitude and navigates to its intended target by venting gas to descend and dropping ballast to ascend.
“We basically took a lot of the innovations that lead to smartphones, global satellite communications, all of the last 20 years of progress in consumer electronics and other things and applied that to balloons,” Dean told me. In the past, the electronics needed to control Windborne’s system would have been too heavy — the balloon wouldn’t have gotten off the ground. But with today’s tiny tech, they can stay aloft for up to 40 days. Eventually, the company aims to recover and reuse at least 80% of its balloons.
The longer airtime allows Windborne to do more with less. While globally there are more than 1,000 conventional weather balloons launched every day, Dean told me, “We collect roughly on the order of 10% or 20% of the data that NOAA collects every day with only 100 launches per month.” In fact, NOAA is a customer of the startup — Windborne already makes millions in revenue selling its weather balloon data to various government agencies.
Now, with a potentially historic hurricane season ramping up, Windborne has the potential to provide the most accurate data on when and where a storm will touch down.
Earlier this year, the company used WeatherMesh to run a case study on Hurricane Ian, the Category 5 storm that hit Florida in September 2022, leading to over 150 fatalities and $112 billion in damages. Using only weather data that was publicly available at the time, the company looked at how accurately its model (had it existed back then) would have tracked the hurricane.
Very accurately, it turns out. Windborne’s predictions aligned neatly with the storm’s actual path, while the National Weather Service’s model was off by hundreds of kilometers. That impressed Khosla Ventures, which led the company’s $15 million Series A funding round earlier this month. “We haven’t seen meaningful innovation in weather since The Weather Channel in the 90s. Yet it’s a $100 billion market that touches essentially every industry,” Sven Strohband, a partner and managing director at Khosla Ventures, told me via email.
With this new funding, Windborne is scaling up its fleet of balloons as it prepares to commercialize. The money will also help Windborne advance its forecasting model, though Dean told me robust data collection is ultimately what will set the company apart. “In any kind of AI industry, whoever has the top benchmark at any given time, it’s going to fluctuate,” Dean said. “What matters is the model plus the unique datasets.”
Unlike Windborne, the tech giants with AI-based weather models — including, most recently, Microsoft — aren’t gathering their own data, instead drawing solely on publicly accessible information from legacy weather agencies.
But these agencies are starting to get into the game, too. The European Centre for Medium-Range Weather Forecasts has already created its own AI-based model, the Artificial Intelligence/Integrated Forecasting System, which it runs in parallel to its traditional model. NOAA, while a bit behind, is also looking to follow suit.
“In the end, we know we can't rely on these big tech companies to just keep developing stuff in good faith to give to us for free,” Kleist told me. Right now, many of the top AI-based weather models are open source. But who knows if that will last? “It's our mission to save lives and property. And we have to figure out how to do some of this development and operationalize it from our side, ourselves,” Kleist said, explaining that NOAA is currently prototyping some of its own AI-based models.
All of these agencies are in the early stages of AI modeling, which is why you likely haven’t noticed weather predictions making a pronounced leap in accuracy as of late. It’s all still considered quite experimental. “Physical models, the pro is we know the underlying assumptions we make. We understand them. We have decades of history of developing them and using them in operational settings,” Kleist told me. AI-based models are much more of a black box, and there’s questions surrounding how well they will perform when it comes to predicting rare weather events, for which there might be little to no historical data for the model to reference.
That hesitation might not last long, though. “To me it’s fairly obvious that most of the forecasts that would actually be used by users in the future will come from machine learning models,” Peter Dueben, head of Earth systems modeling at the European Centre for Medium Range Weather Forecasting, told me. “If you just want to get the weather forecast for the temperature in California tomorrow, then the machine learning model is typically the better choice,” he added.
That increased accuracy is going to matter a lot, not just for the average weather watcher, but also for specific industries and interest groups for whom precise predictions are paramount. “We can tailor the actual models to particular sectors, whether it's agriculture, energy, transportation,” Kleist told me, “and come up with information that's going to be at a very granular, specific level to a particular interest.” Think grid operators or renewable power generators who need to forecast demand or farmers trying to figure out the best time to irrigate their fields or harvest crops.
A major (and perhaps surprising) reason this type of customization is so easy is because once AI-based weather models are trained, they’re actually orders of magnitude cheaper and less computationally intensive to run than traditional models. All of this means, Kleist told me, that AI-based weather models are “going to be fundamentally foundational for what we do in the future, and will open up avenues to things we couldn't have imagined using our current physical-based modeling.”
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Europeans have enjoyed it for years. Now, through careful state interventions and creative salesmanship from startups, Americans are close to having their turn.
For U.S. consumers, going solar is usually a major undertaking, involving tens of thousands of dollars, months of logistics, a slew of financing options, and ever-changing incentives.
But in Germany, upwards of a million customers — homeowners and renters alike — are simply plugging in small, affordable solar arrays to standard power outlets. These small systems are, by law, 800 watts or less, a fraction of the size of a typical rooftop solar system in the U.S. Often called “balcony solar,” these panels can live essentially anywhere with sufficient sunlight: on balconies or patios, or mounted on exterior walls or flat rooftops.
But while governments across the EU have simplified regulations to make installation a quick, DIY process, and utility approval little more than a formality — unleashing a wave of consumer demand in the process — the U.S. has so far failed to follow suit. Here, utility regulations prohibit customers from feeding power back into the grid without a formal interconnection agreement, a process that involves lots of time and paperwork.
Utilities in the U.S. want to account for all electricity sources on the grid, since theoretically, even small plug-in systems could have a cumulative impact on local voltage and power quality, whereas in Germany, for example, this is less of a concern. There, plug-in solar-specific policy caps these systems’ generating capacity, and the grid and metering infrastructure has been more extensively modernized to handle distributed energy generation.
Now, however, there are a number of domestic plug-in solar startups finding creative ways to navigate the constraints of the U.S. market. One of them, the nonprofit Bright Saver, announced on Wednesday that it’s raised $500,000 in new funding from TrueVentures.org and a handful of individual backers. The company gets around power export regulations by selling panels with very low wattage. “So we’re talking 200- or 220-watt systems that never backfeed to the grid, because we think close to every typical household will consume that electricity immediately, simply with the refrigerator,” Cora Stryker, the company’s co-founder, told me.
The San Francisco-based startup has sold a couple dozen systems already and has a waitlist of about 1,500 people, Stryker said. So far, she told me, the majority of this “early adoption crowd” is mainly interested in reducing their own emissions. “We think that’ll change over time,” she said. “The mass adoption in Germany has been driven not by that climate-conscious crowd, but really people who want to save money. “
The main drawback to Bright Saver’s approach, however, is also what makes it possible in the first place: the panels’ incredibly small size, which can’t come close to covering a home’s full power needs. So while the upfront cost of a 200-watt panel is small — $399 at the moment — a customer’s energy savings will also be tiny — potentially on the order of just a few bucks per month. Depending on the location, the savings will eclipse the total cost in about five to 10 years, Stryker told me.
That might not be enticing enough to convince a critical mass of customers to jump onboard the small-scale solar train. But Stryker thinks that getting these products out into the world will help catalyze the type of curiosity and interest that can dovetail into policy change. “Selling product in the next year or two is a small revenue stream for us, but it’s also our theory of change,” she told me. “These need to get out there in order for people to know they even exist.”
Much of Bright Saver’s work involves advocating for easing plug-in solar regulations, which is already starting to happen, bit by bit. In March, the Utah state legislature unanimously passed a bill creating a new category for “small portable solar generation devices” under 1,200 watts, exempting them from interconnection requirements. Stryker told me that Utah’s governor was inspired to introduce the bill after reading a story in The New York Times about balcony solar’s success in Germany.
Now more states, including Vermont, Maryland, and Pennsylvania, are expressing interest in similar legislation. If just a few more get onboard, Stryker told me that would be a critical tipping point. “We’ve had conversations with manufacturers and investors who tell us straight up, they’re not coming to the U.S. market because they see only one state where they’re not going to run into these regulatory concerns,” she said. “They tell us privately, five to seven more states and they’re in. So that’s a key threshold for us.”
But one veteran of the plug-in solar market, Craftstrom, isn’t betting on this happening. The company has been selling 400- to 800-watt systems in Europe since 2017, and expanded into the U.S. a few years later, targeting markets where electricity prices are highest, like California and the Northeast. To deal with domestic regulations, the company patented a new type of meter to be placed inside electric panels that blocks excess power from flowing back into the grid. This prevention mechanism also allows the company to sell larger systems — up to 2,000 watts — in the U.S..
Craftstrom’s chief revenue officer, Ken Hutchings, thinks this type of system is critical for grid safety in the U.S., where distribution networks tend to be older and less standardized than in Europe, and not necessarily built for two-way power flow. This opens up utilities to a good deal of legal liability in the case of equipment failures.
While Hutchings wouldn’t necessarily be surprised to see other states following Utah’s lead, he’s skeptical that the U.S. will become a haven for plug-in solar anytime soon — or even that it’s a good idea. “There’s no risk to one or two guys pushing power back into the grid,” he told me. “But when you have thousands and thousands of people doing it, tens of thousands, and the electric company is not sure who’s doing it, I think that’s where the issue lies.”
Thus far, Craftstrom has sold about 4,000 units in the U.S., with about 500 of those orders coming in the past month alone, Hutchings told me. He attributed the sudden uptick largely to a rush of customers trying to qualify for home energy efficiency tax credits — which he said Craftstrom’s systems are eligible for — before they expire at year’s end.
Craftstrom’s domestic prices are still more expensive than what its own customers in Europe can expect to pay for similar systems due to the extra hardware costs that come along with the specialized meters, as well as the fact that installing these products is not a DIY operation. That means Utah customers should now enjoy the same price relief, since the new state law lifts the grid restrictions that the rest of the U.S. faces. These days, Craftstrom’s more complex hardware plus the cost of labor “just about doubles the cost from what you’re able to get in Utah,” Stryker told me.
Bright Saver sold Craftstrom’s systems when it first started out earlier this year, but chose to discontinue this offering as it “didn’t serve our vision of making this accessible to everyone through cost and self-installation,” Stryker told me. Instead, the organization is focusing on policy changes that will make cheap self-install systems in the 800-watt range feasible in more states. And that means getting legislators onboard with some degree of deregulation, something Stryker acknowledges “has often been a dirty word” in the environmental movement.
“In this case, we need these regulations to get out of the way. They’re outdated. They’re artifacts,” she told me, referring to the requirement that small plug-in systems sign utility interconnection agreements. “I see it as a purple narrative, one that can appeal to values across the political spectrum — energy independence, energy affordability, renters’ rights.”
Of course, Stryker isn’t advocating for complete anarchy in the space. Grid stability is still a concern, and she said that Bright Saver is involved in discussions with regulators and standard-setting bodies to determine acceptable wattage thresholds. Countries that have embraced balcony solar in Europe have “impeccable” safety records, Stryker told me, enabling Germany to raise its wattage limit from 600 to 800 watts at the beginning of last year.
There are still some logistics to work out though. As the recent Utah law is written, plug-in solar arrays must comply with product standards from Underwriters Laboratories, a safety certification body. And while this organization has standards covering the individual components of plug-in solar systems, it has yet to create a systems-level standard. Depending on whom you ask, that might mean all domestic companies in the space are operating in a bit of a regulatory gray area at the moment.
Stryker told me she expects these system-wide standards to be released soon though, ideally in tandem with more bills like the one passed in Utah. “We think it’s a no-brainer.”
On Alaska’s permitting overhaul, HALEU winners, and Heatmap’s Climate 101
Current conditions: Kansas, Oklahoma, and Arkansas brace for up to a foot of rain • Tropical Storm Juliette, still located well west of Mexico, is moving northward and bringing rain to parts of Southern California • Heat and dryness are raising the risk of wildfire in South Africa.
The Trump administration has started the process to roll back logging protections from more than 44 million acres of national forest land. On Wednesday, U.S. Secretary of Agriculture Brooke Rollins proposed undoing a 25-year-old rule that banned building roads or harvesting timber on federally controlled forest land, much of which is located in Alaska. “Today marks a critical step forward in President Trump’s commitment to restoring local decision-making to federal land managers to empower them to do what’s necessary to protect America’s forests and communities from devastating destruction from fires,” Rollins said in a statement. “This administration is dedicated to removing burdensome, outdated, one-size-fits-all regulations that not only put people and livelihoods at risk but also stifle economic growth in rural America.”
Environmental groups slammed the proposal for jeopardizing wildlife habitats and putting waterways at risk. “Communities depend on clear water filtered by roadless areas, animals depend on the unfragmented habitat that can only exist where there are no roads, and anglers depend on clean water in the streams where trout and salmon swim,” Ellen Montgomery, the director of Environment America’s great outdoors campaign, said in a press release. “We cannot let these essential forests be carved up by roads, obliterated by chainsaws, and contaminated by mines.”
Heatmap’s new Climate 101 series aims, as Heatmap deputy editor Jillian Goodman explained, to be “a primer on some of the key technologies of the energy transition.” That includes “everything from what makes silicon a perfect material for solar panels (and computer chips), to what’s going on inside a lithium-ion battery, to the difference between advanced and enhanced geothermal.”
This might be especially helpful for those still trying to find their way into the climate conversation, but we hope there’s something here for everyone. For instance, did you know that contemporary readers might have understood Don Quixote’s “tilting at windmills” to be an expression of NIMBYism? Well, now you do!
The federal Permitting Council signed a first-of-a-kind memorandum of understanding to work together with Alaska’s government to streamline permitting on critical infrastructure projects across the state. First established in 2015, the agency was designed to improve transparency and speed up the greenlighting of infrastructure approvals. But it had yet to forge such a close pact with an individual state. “Our team is ready to work with Governor Dunleavy to bring Alaska back into the energy spotlight, ending the neglect of the Biden Administration and bringing Alaska’s incredible natural resources to the rest of the world,” Emily Domenech, the Permitting Council’s executive director, said in a statement.
Domenech — a former staffer for House Speakers Kevin McCarthy and Mike Johnson who went on to serve as a senior vice president at Boundary Stone, a firm founded by alumni of the Obama-era Department of Energy — acted as something of a Republican sage for the clean energy industry. In an interview with Heatmap’s Matthew Zeitlin after last November’s election, she urged the industry to forge closer relationships with members of the current congressional majority. “If you ask Republicans to be for or against the IRA as a whole, they’ll be against it,” Domenech said, “But Republicans think about energy as a regional issue. So instead of forcing this one size fits all approach, IRA advocates would be smart to give people room to support only the policies that make the most sense for their state or region.”
The Department of Energy selected another three companies to receive a special kind of nuclear fuel from its growing stockpile. HALEU — pronounced HAY-loo, an acronym for high assay low enriched uranium — is a reactor fuel enriched up to four times as much as traditional reactor fuel. The fuel is needed for all kinds of novel reactor designs, particularly those that use coolants other than water. Until recently, however, Russia’s state-owned Rosatom had enjoyed a virtual monopoly over its global supply. The Biden administration set aside billions for HALEU production. In April, the Trump administration selected five companies to receive some of the government-procured supply, including Westinghouse, Bill Gates’ TerraPower, and the Google-backed Kairos Power. Now the agency has picked another three:
Two firefighters battling the Bear Gulch fire on Washington’s Olympic Peninsula were arrested by federal law enforcement Wednesday. The reason for the arrests is unclear, according to the Seattle Times. Over three hours, federal agents from Border Patrol carried out an “operation on the fire,” demanding identification from members of two private contractor crews who were among the 400 firefighters battling Washington state’s largest active blaze. The Incident Management Team from the National Interagency Fire Center suggested that the action did not interfere with the efforts to tamp down the flames.
The American West is primed for wildfires right now. Following a lull in June and July, Heatmap’s Jeva Lange wrote that “the forecast for the Pacific Northwest for ‘Dirty August’ and ‘Snaptember,’ historically the two worst months of the year in the region for wildfires,” was full of warning signs, including low precipitation and abnormally high temperatures.
Living, gnawing weedwackers.Vesper Energy
The 1.36 million solar panels at Vesper Energy’s Hornet Solar farm in Swisher County, Texas, one of the United States' largest single-phase solar projects, were overgrown with vegetation. So naturally, the company brought in sheep. More than 2,000 white, wooly ovines arrived this month and were allowed to roam the facility’s six square miles. “As Texas continues to lead the nation in solar energy growth, solar grazing highlights how innovation can support rural economies, preserve farmland, and strengthen the state’s reliable energy future,” Vesper said.
Here at Heatmap, we write a lot about decarbonization — that is, the process of transitioning the global economy away from fossil fuels and toward long-term sustainable technologies for generating energy. What we don’t usually write about is what those technologies actually do. Sure, solar panels convert energy from the sun into electricity — but how, exactly? Why do wind turbines have to be that tall? What’s the difference between carbon capture, carbon offsets, and carbon removal, and why does it matter?
So today, we’re bringing you Climate 101, a primer on some of the key technologies of the energy transition. In this series, we’ll cover everything from what makes silicon a perfect material for solar panels (and computer chips), to what’s going on inside a lithium-ion battery, to the difference between advanced and enhanced geothermal.
There’s something here for everyone, whether you’re already an industry expert or merely climate curious. For instance, did you know that contemporary 17th century readers might have understood Don Quixote’s famous “tilting at windmills” to be an expression of NIMYBism? I sure didn’t! But I do now that I’ve read Jeva Lange’s 101 guide to wind energy.
That said, I’d like to extend an especial welcome to those who’ve come here feeling lost in the climate conversation and looking for a way to make sense of it. All of us at Heatmap have been there at some point or another, and we know how confusing — even scary — it can be. The constant drumbeat of news about heatwaves and floods and net-zero this and parts per million that is a lot to take in. We hope this information will help you start to see the bigger picture — because the sooner you do, the sooner you can join the transition, yourself.
Without further ado, here’s your Climate 101 syllabus:
Once you feel ready to go deeper, here are some more Heatmap stories to check out: